Hypocretin (Hcrt, also called orexin) neurons have been implicated in the pathology underlying narcolepsy. We found that human narcoleptics, on average, have a 90% loss of Hcrt cells. Although the focus of most work has been on Hcrt's arousal function and its purported relation to food intake, our recent work has shown that activity of these cells is not simply related to arousal, food intake or unearned reward. Rather it is specifically linked to positive emotion and arousal during work for rewards (Progress Report). Hcrt neurons have major projections to dopamine neurons and appear to share and perhaps mediate aspects of dopamine's involvement in reward. Anecdotal data has long suggested that Hcrt deficient human narcoleptics are resistant to drug addiction. This finding has now been replicated in animal studies. Hcrt and MCH (melanin concentrating hormone) neurons are intermixed and appear to have reciprocal discharge profiles across the sleep-wake cycle. The number of Hcrt cells that we see in normal humans ranges from 51,000-83,000 based on the 15 normal brains we have counted. However, we recently received a normal brain that contained 114,000 Hcrt cells, 37% greater than the highest number of Hcrt cells (83,000) and 50% greater than the mean number (76,000), seen in controls. We also found that the size of Hcrt neurons in this individual was significantly smaller than that of controls. Further investigation revealed that this individual wa a former heroin addict, who had not taken addictive drugs for >10 years before his death. This suggested that opiate administration might alter the Hcrt system and that this alteration might be related to the craving experienced by former addicts. In pilot studies for this proposal, we then found that Hcrt cells in mice sacrificed after a 3 day period of continuous morphine administration were of normal number, but had reduced neuronal volume and altered morphology. This was not a subtle effect. Shrinkage averaged 23% in area after only 3 days of morphine administration. Furthermore, when we administered morphine for 7 and 14 days, we saw a substantial increase in Hcrt cell number. Such changes have never been reported in Hcrt neurons. In the same study, we found that morphine administration greatly decreased the number of identifiable MCH cells, a cell group not previously implicated in addiction. In contrast to Hcrt cells, the size and morphology of MCH cells was normal. The goal of the current proposal is to determine the nature and time course of morphological changes in Hcrt and MCH cells produced by short term morphine administration, by morphine addiction and by morphine withdrawal. We hypothesize that some of these changes are irreversible. We will study the behavioral correlates of these anatomical changes. These might include alterations in performance of positive vs. negatively reinforced tasks and modifications of sleep. We will secure additional human addict brains to replicate our initial observation of Hcrt cell number increase. Establishing that the properties of Hcrt cells can be altered in this proof of principal study might not only lead to a better understanding of drug addiction, but also lead to techniques to permanently and therapeutically alter and perhaps augment Hcrt function in disorders with Hcrt dysfunction including narcolepsy and depression.